An active monitoring system for real-time face-tracking based on mobile sensors

Surveillance systems frequently use fixed or semimobile cameras. However, in many cases, the use of intelligent mobile sensors is preferrable over fixed sensors, because the system configuration can be modified according to particular environmental conditions or adapted to compensate for one or more malfunctioning sensors. This paper proposes a real-time surveillance system based on a mobile sensor. Using an Android smartphone and a face-tracking algorithm, the system can move autonomously to track the human face with the longest presence in the video field. In addition, the system can be connected to a computer performing face-recognition through a wireless connection provided by the smartphone. This way the mobile sensor can track a determined human face. The paper provides some experimental results to validate system performance.